Open Source cheap ventilator V2. Design optimisation

  • Status: Closed
  • Prize: $200
  • Entries Received: 13
  • Winner: sami3008

Contest Brief

This is a follow on from the first round. Now we need to collaboratively combine and improve the design. Like the first round I will award two or more second places of $100 but this is a token amount as we are trying to save lives here. If I get any donations (cheapventilator g. mail) then I will increase the prizes but I have already spent a lot of my own money on pumps, fans, valves, gears, sensors, filters... and the previous prizes.

This is a collaborative competition, please share your good ideas but don't waste people's time with too many weak ideas. As in the last competition I valued sharing ideas and sources of components, suppliers etc. I will add my own comments and updates regularly and I will share my comments with each collaborator with the whole group.
The good news is we may be getting some sponsorship so I can perhaps offer more than token prizes. I will give you good feedback where I can and hope this is a rewarding experience for those participating. Keep safe and I hope we never need to use these :-)
Link to all the designs so far: [login to view URL]

Suggest which single design we use as the basis going forward and which ideas from others we include then combine, simplify, optimise for 3D printing, buying standard components etc.

A related project to optimise the valve:
https://www.freelancer.co.uk/projects/mechanical-engineering/Optimise-geometry-rotating-valves-for

Recommended Skills

Employer Feedback

“Again Mohamed beat other good engineers by coming up with very creative solutions, working quickly and with good attention to detail, and he was able to foresee problems and offer solutions to mistakes I had made and issues we had not anticipated. Thanks again for the good engineering and dedicated efforts.”

Profile image derekrwhite3, Canada.

Top entries from this contest

View More Entries

Public Clarification Board

  • derekrwhite3
    Contest Holder
    • 1 month ago

    Many thanks to Mohamed SM and Ahmad QZ for some really excellent work! I have selected Mohamed as the winner due to his clever adjustable inspiratory pause blade with its own markings and a barrier to stop air leaks and his speed at solving problems and offering solutions. Ahmad will get second place and we will use some of his good ideas in the combined first prototype. Thanks to you both!!

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    To allow easier injection moulding the front cover for the gearbox must be a separate part. I suggest a small separate part (tube with central wall) fits into a round recess to transmit air through gearbox and then the valve pack presses against that. Another tube part then screws onto the front cover (perhaps will rubber seal which pushes against valve pack.)

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Ahmad #9 : Thanks Ahmad. Nice simple design. 1) I suggest we have the HEPA filter slide into the bottom through a door (or inside a drawer) in the side opposite the air in. The inlet can be lowered to near the bottom and feed under the filter (add a horizontal wall to direct flow if needed.) The air passes up through the filter and then into the rotating valve. 2) The low pressure box and vacuum exit box can be only 2/3 depth of the total height, The bottom 1/3 of those two boxes can be a feed from the high p side across to a hole under the fan to pressurize the gear case and supply air to the fan. For the fan speed valve, why not have a rotating blade valve sitting in a slot or in a U-shaped guide, tab up through the case top. rotate to open or close. 3) what filter can we fit into there. choose the biggest that is cheap, easy to find and can fit without making the device too big and expensive?

    • 1 month ago
  • qhaliffzuber
    qhaliffzuber
    • 1 month ago

    Mohamed Sami, which HEPA filter are you using and the dimensions?

    • 1 month ago
    1. sami3008
      sami3008
      • 1 month ago

      Hi Ahmed, I am using the V3 filter (80*40*15), one of the filter suggeted by Derek

      • 1 month ago
    2. derekrwhite3
      Contest Holder
      • 1 month ago

      I will try to find flow data for that filter. We may need to use two or a bigger filter. Did you use the Bob Sensig spreadsheet to check the flow rate of a typical vacuum (50-100 cfm?)

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Mo SM #8 ; Thanks Mohamed, MY comments 1) Vacuum gauge not needed. 2) What filter are you using, if any? 3) Lets move the bypass and fan speed valves to the top. please send the open top view without flow arrows so I can draw in some changes. Thanks. Now that I am home I can test some fans for speed, torque, and maybe we need fewer gears. Thanks!

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Here are the dimensions of those Miele filters in the spreadsheet his HEPA H12 filter soaks up thanks to an extremely high filtration even the smallest pests such as pollen, spores and household dust on that cause allergies and respiratory. So, this product is ideal for allergy and hay fever sufferers. Thanks to a special active carbon layer are unpleasant odours from the air. For optimal filter performance and mild after 2 years be replaced with another of a different HEPA (High efficient particle arrester) filters every six months.
    Suitable for use with the following vacuum cleaners:
    Miele Parkett & Co 5000
    Suitable for the Miele ranges:
    S 4210 – S 4782
    S 5210 – S 5981
    S 6240 – S 6760
    S 8000 – S 8930
    Product Details:
    Length: 180 mm
    Width: 80 mm
    Height: 28 mm

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Some info for the filter spreadsheet: Bob Senzig uploaded a file.
    Filter design is key for Mask and PAPR effectiveness. From other data HEPA is clearly an option but the filter must be big enough or the air speed through the filter will rise and efficiency will drop. Commercial 6CFM system Use around 400 sq in of total filtration. To achieve this. They use a pleated filter. Face masks often do not because the human breathing is around a 0.35 CFM pump.(10-20 breaths/min with 0.5liter Tidal volume) A facemask should have around 20 sq in (7inch diameter). Here a worksheet to calculate filter area and look up the filter efficiency given your blower capability. This collects a lot of info and represents an attempt to explain a little about filters. I don't design them for a living, but I think I got this right please give me feedback.
    The attached file allows the user to enter the dimensions of the filter and the number of folds in the filter to calculate the effective filter sq inch area.

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    added a useful calculator for small HEPA filters into files:

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Panda PEEP is now gen 3 with steel spring https://www.thingiverse.com/thing:4278384?fbclid=IwAR2TUneWMEv2oZ_T6lzEp8-EOFs9Oxo2EiqdvT5iC3akj2uLy9acWAklcYg

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    We can effectively have a pause at the end of exhale by increasing the exhale duration but we can't have a pause between inhale and exhale unless the pie blocks the flow for a bit (by being bigger than the hole in the rotating valve. But I think no pause is needed. One worry I have is than even with 0.5mm steel disks, the air can short cut between the two sides through the small 0.5mm gap. Can you suggest a solution? Maybe my idea of the teflon sealing disk would work? Maybe repeat one of the disks so that the central one is sandwiched to stop leaks?

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    low pressure PEEP must go out via filter otherwise we don't vent any air. High p side safety valve will be same design but will be circulated. For the new file showing the cheap thermometer/hygrometer, we should see if there is space to add it as an option (glue it on with RTV silicone and have a hole drilled into the air flow. We only need a hole for air on the right side on the back when looking from the front (side with two holes.)

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    For anyone working on the rear case? Remember the holes for 63mm OD pressure gauges with back mounting threaded connection (probably 1/8NPT.) they should not interfere with any valves etc. Please hurry as 3 days left and I’m not extending. Thanks derek

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    For anyone intending to participate, please show your designs as soon as possible so that we can add your ideas into the discussion and improve the evolving design. Designs submitted at the last minute do not help anyone else so are not collaborative in the spirit of saving lives. Thanks

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    port walls are at 210 degrees and 330 degrees, blade pies are centered on the walls and will be at least 90 degree "pies". I have asked an expert for the pause between inhale/exhale as the pies can be >90 degrees perhaps depending on desired pause duration. if you find any info on pause between inhale and exhale please share as it will determine disk pie size

    • 1 month ago
    1. derekrwhite3
      Contest Holder
      • 1 month ago

      This comment is assuming we have only two disks. Four would give us unlimited control but perhaps not needed. Degrees measured from the top and going clockwise as per convention.

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    The graphs I've seen of pressure during BiPAP suggest no pause between breaths or holding breath. Let's stick to 90 degree blade pies. Next job is to decide if complex rear housing can better hold passages and valves than PVC pipes etc. Not sure which is easier to build. Since we have to make custom front section for gears, valves etc. maybe a back section can have all the passage plus a vacuum HEPA filter option if a fan is used instead of a vacuum cleaner (an extra filter is always good if several patients/devices are powered by the same vacuum.

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    My suggestion if we use two blades is now in files. We could have teflon disks between the valve disks and/or between them. Sketch shows the 120 degree inhale and 240 degree exhale ports

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    I was thinking about the valve blades-- they probably only need to be 0.5mm thick and the thinner they are, the less air will leak between the high a low pressure sides. We may improve the sealing between sides and reduce friction by having a thin teflon spacer with the middle "bridge seal" which has a tab at the bottom in a slot to stop it rotating. Also, unless we really need to adjust the pause at the end of inhale and the end of exhale, then we only need two blades, not 4.

    • 1 month ago
  • qhaliffzuber
    qhaliffzuber
    • 1 month ago

    A link to lock bearing for 4mm shaft. https://www.bearingboys.co.uk/Solid--Unsplit/CABU04Z-4mm-Shaft-Collar-SolidUnsplit-40256-p

    • 1 month ago
    1. sami3008
      sami3008
      • 1 month ago

      Thanks Ahmad

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Thanks to Mohamed SM and Ahmad Q for the updated files. Please all look at these and comment, improve, change etc. Added to files

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Let's do the valve control blades out the sides-- inhale controls on the high pressure side and exhale on the low pressure side. Seems logical and easier to see and reach. We can also have markings (embossed or sticker) showing numbered degree markings as these will aid the adjustment

    • 1 month ago
    1. sami3008
      sami3008
      • 1 month ago

      Hi everyone, I was working on the thin blades. I found the solution to integrate them in the main body and make them adjustable from the outside. There is no space between the blades to avoid leaks. Please look at my last entry

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    A good way to have some audible warning or feedback is a whistle, air buzzer, or air bell on the outlets of the pressure valves. While the high pressure side vent whistle volume and pitch will indicate too much air pressure, the PEEP valves intermittent whistle will notify the device is working and the BPM. Not sure if a whistle inside the high pressure return tube will be audible (or even work inside a vacuum?) Can anyone find a low-flow whistle off the shelf? Kids toy whistle? Whistle design link?

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Mo SM's files are now on google docs (https://bit.ly/2ReyRLa) and some extra renderings with flow in the files section. This is really great I I suggest we work to improve this design. A couple of thoughts: 1) extra pipes and valve between high pressure side and low pressure side, we need slight positive pressure PEEP and that pressure can come from the high pressure bleed, 2) extra valve between PEEP and vacuum so that we can have manually activated negative pressure for cough response, airway clearing, 3) the original labyrinth was to separate the O2 from the high flow air (where some of the O2 would be lost/wasted. Is this a real concern or will the O2 be blended well and only escape with exhale via PEEP to outside? 4) once we add all of these things, is it still easier to assemble from std PVC parts or should we put it into the complex housing from previous designs? What do you think? Might be easier to have all that complex flow made by the housing and not external std parts?

    • 1 month ago
    1. qhaliffzuber
      qhaliffzuber
      • 1 month ago

      If we want to go back to the original housing, might I suggest making it modular? The housing for gears, valves, etc is different for the housing of filter and housing of humidifier. One advantage of this is maybe to remove the use of humidifier should it be not needed?

      • 1 month ago
    2. derekrwhite3
      Contest Holder
      • 1 month ago

      Thanks for clarifying Mo. The knurled knob behind the rotating valve is a needle valve which opens a passage in the dividing wall between high pressure side and low pressure side. Good simple solution to accurately control the bleed of air to the low pressure side and ensure sufficient air to maintain the PEEP pressure but not so much that we vent too much air/O2 continuously out the PEEP valve. Might be a good idea to just have a std O2 connector there and a shutoff valve for those not using O2 and/humidifier.

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    For prototypes we can use standard gears but may have custom 70/12 or 60/10 gears made with our quarter pie hole for volume production. Some gear CAD files at https://khkgears2.net/catalog2/DS0.8-70

    • 1 month ago
    1. sami3008
      sami3008
      • 1 month ago

      For prototypes also, to create air passages, we can use standard fittings and tubings instead of PVC or 3 printed parts :
      https://www.industrialspec.com/shop/medical-component-products/medical-fittings/antimicrobial-fittings.html
      https://www.industrialspec.com/shop/medical-component-products/medical-grade-tubing.html

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    good link from Mo SM. Can we get these or similar? Might be in short supply now: https://www.armstrongmedical.net/product/peep-valves/

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    "we do want the steel disks to be independently adjusted from outside the housing right?" yes. I thought the long tabs can be bent to give small handles and maybe rounded plastic knob to protect skin from sharp edges. Perhaps inhale side two facing forwards (one longer than the other) and exhale side two facing backwards.

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Only £5 for these thermometer/hygrometers. 80mm x 50mm x 10mm and they have the sensors on the back (small hole) so these can be mounted easily with RTV and a small hole drilled in line with sensor hole. https://www.amazon.co.uk/dp/B07YY99MQD/

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    I also think we need a bigger surface area filter. The small low-flow mask filters are both hard to source now and they have a very small surface area. Lets go back to a pleated robot vacuum or similar small vacuum cleaner HEPA filter (big surface area, easier to source, cheaper.) The bleed off of exhaled air via the PEEP valve can have a small filter disc as flows are much lower.

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Ahmad will try to get a test sample of his PEEP valve 3D printed. My comments were: Make the housing have support for 3M or Molex disk, cloth, toilet paper, anything people may use to filter the exhaled air. Base the first model on the pen spring and I suggest a tube on the disk and a pin on the adjuster so the the pin slides into the tube with the spring outside the tube (keeps tube central). Ahmad please send the CAD file so others can try them out of improve the design. I'll add it to the files. Thanks!

    • 1 month ago
  • sami3008
    sami3008
    • 1 month ago

    Hi every one, I am sharing with you tis updated design. I preserved the old part including the gears and the rotating valve and I designed new air passages.
    For the air passage we can use PVC or ABS pipes with accessories (elbows, Tees, etc.) or we can make them using 3D print.
    Please find below the links to the standard parts :
    PEEP valve : https://www.armstrongmedical.net/product/peep-valves/
    Bacterial-Viral Filter : https://www.mvapmed.com/clear-bacterial-viral-filter.html#product-details-tab-specification
    Humidifier : https://www.airtechmedical.com/product/medical-oxygen-regulator-humidifier-bottle/
    Knife gate valve : https://www.praher-plastics.com/product/praher-knife-gate-valve/

    • 1 month ago
    1. qhaliffzuber
      qhaliffzuber
      • 1 month ago

      Nice design Sami! what dimension of gears are you working on? Is that following the standard catalog that you shared the other day?

      • 1 month ago
    2. sami3008
      sami3008
      • 1 month ago

      @Ahmad Qhaliff Z. Yes I am using standard gears from the catalog I shared with you. I used three reduction 12:70 * 12:70 * 12:70 with a 0.8 module. The total ration is nearly 1:198

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    I wonder if it would not be better to put the filter underneath. Typical CPAP tubes are 22mm OD so our passages could be similar cross section area. if we use 60-70 sqmm as our flow paths inside then the size, cost and time to 3D print come down.

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    For Ahmad's adjustable pressure valve, google search suggested ball point pen springs are about 250N/m which is about 25g/mm I think so it might work. Can you check my maths? We want to be able to adjust the range from 2-3 cmH2O up to 40cmH2O with the travel given by the threaded adjustment.

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Can we start to make changes ASAP please. How will we mount the gears and valves for the alternating flow? My suggestion of a removable side plate? What gears? How will we best flow air to the fan, through the gearbox or overhead through the lid? First decent proposal needed ASAP. I have some back entry threaded pressure gauges coming for testing so they will go on the top as well. The sliding gate valves are simple, can we size them based on needed flow? Can anyone suggest a model or simulation tool to size the valve sizes? can someone adapt the two Pressure valves for cone springs as per the CAD files I suggested?

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    Ahmad's PEEP:
    Great, is this a std HEPA filter disk like used on mask (eg. 3M 2067 2091 etc.?) Add short barbed hose section so we can vent to vacuum inlet if needed for more filtering of high pressure side. Low pressure side can be where we lose air. We can use same valve as a backup safety set and fixed to 40cmH2O

    • 1 month ago
    1. qhaliffzuber
      qhaliffzuber
      • 1 month ago

      https://www.safetycompany.com/respirators/reusable-respirators/moldex-7940-p100-particulate-filter-disk-for-7000-series-half-mask-2-each/. Is this type doable? I have not based the HEPA filter on anything before this

      • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    OK, please look at my design suggestions and assume 1mm thick stainless steel valve disks and a 40mm diameter gear-- 60 tooth/10tooth, all gears the same. Need to sort out geometry-- perhaps final valve gear against the housing inside wall (for air seal, perhaps with 0.2mm teflon sheet to slide on both sides), then the other gears have spacer bosses on the housing to offset them. How do we easily get air to the fan (around or over?)

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    for a more compact design maybe use this spring with a 1.56cm2 valve disk https://www.assocspring.co.uk/conical-springs/A0600-029-0625-S

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    To limit the pressure to around 40cmH2O (with approx a 1sq.cm disk we need something like this: https://www.assocspring.co.uk/conical-springs/A0600-029-1250-S

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    heres one calculator https://www.f650gs.crossroadz.com.au/CalcSpring.html . Perhaps 0.8mm wire 10mm long, 10mm diameter, pushing onto about 1 cm2 disk will give about 66g/cm2

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    For out two pressure relief valves: if ballpoint spring rate is too low pressure maybe a conical spring, lts available for cheap like these. maybe there is an online spring rate calculator based on diameter wire thickness material https://www.ebay.co.uk/itm/15mmx6mmx26mm-Metal-Taper-Conical-Compression-Spring-Silver-Tone-30pcs/362000816165?hash=item5448ee1825:g:v9gAAOSwfkxeeYYu

    • 1 month ago
  • derekrwhite3
    Contest Holder
    • 1 month ago

    We need to keep in mind that the rotating disk valve and the pack of four thin adjustment disks must be squeezed together so must be constrained on each side by the main housing and the cover, both with a central piece to support the shaft and also to separate the inhale and exhale sides. We may use an adapter to plug into the mouthpiece inlet/outlet which will keep the two air flows separated for two hoses.

    • 1 month ago

Show more comments

How to get started with contests

  • Post your contest

    Post Your Contest Quick and easy

  • Get tons of entries

    Get Tons of Entries From around the world

  • Award the best entry

    Award the best entry Download the files - Easy!

Post a Contest Now or Join us Today!